Phenol is an important product in the chemical industry and is useful in, for example, the production of phenolic resins, bisphenol A, ε-caprolactam, adipic acid, and plasticizers.
Currently, the most common route for the production of phenol is the Hock process via cumene. This is a three-step process in which the first step involves alkylation of benzene with propylene in the presence of an acidic catalyst to produce cumene. The second step is oxidation, preferably aerobic oxidation, of the cumene to the corresponding cumene hydroperoxide. The third step is the cleavage of the cumene hydroperoxide generally in the presence of a sulfuric acid catalyst into equimolar amounts of phenol and acetone, a co-product.
It is known that phenol and cyclohexanone can be co-produced by a variation of the Hock process in which cyclohexylbenzene is oxidized to obtain cyclohexylbenzene hydroperoxide and the hydroperoxide is decomposed in the presence of an acid catalyst to the desired phenol and cyclohexanone. Although various methods are available for the production of cyclohexylbenzene, a preferred route is disclosed in U.S. Pat. No. 6,037,513, which discloses that cyclohexylbenzene can be produced by contacting benzene with hydrogen in the presence of a bifunctional catalyst comprising a molecular sieve of the MCM-22 family and at least one hydrogenation metal selected from palladium, ruthenium, nickel, cobalt and mixtures thereof. The '513 patent also discloses that the resultant cyclohexylbenzene can be oxidized to the corresponding hydroperoxide which is then decomposed to the desired phenol and cyclohexanone co-product.
In the cumene-based Hock process, dilute cumene hydroperoxide from the cumene oxidation step is first concentrated to greater than 80 wt % by removing unreacted cumene under vacuum, and the resultant concentrate is then sent to the cleavage reactor. In addition to the hazards associated with handling concentrated hydroperoxide, the cleavage poses safety concerns due to the rapid and highly exothermic nature of the reaction. Further, significant amounts of by-products may be generated from the concentrated oxidation products. In practice, therefore, the concentrated cumene hydroperoxide is often diluted with solvents, such as acetone, in order to better manage the heat of reaction and to control by-product formation. For example, U.S. Pat. No. 5,254,751 discloses a method of producing phenol and acetone by decomposing cumene hydroperoxide in a non-isothermal manner in the presence of excess acetone whereby the molar ratio of acetone to phenol in a decomposition reactor is from about 1.1:1 to 1.5:1.
In producing phenol from cyclohexylbenzene, the problems are different. Firstly, oxidation of cyclohexylbenzene to cyclohexylbenzene hydroperoxide is much more difficult than oxidation of cumene and requires elevated temperatures and the use of a catalyst, such as N-hydroxyphthalimide (NHPI). As a result, the cyclohexylbenzene oxidation effluent is also generally at elevated temperatures so that cooling this stream back to ambient temperature would incur additional operating cost. Also, in view of the high boiling point of cyclohexylbenzene, concentration of the cyclohexylbenzene hydroperoxide by evaporation of the unreacted cyclohexylbenzene is difficult and can lead to unwanted decomposition of the hydroperoxide. Thus, with cyclohexylbenzene hydroperoxide cleavage, the feed contains about 80 wt % hydrocarbon and the products contain only about 20 wt % polar components, which limits sulfuric acid solubility and the cleavage rate. In addition, the cleavage chemistry for cyclohexylbenzene hydroperoxide is much more complicated than that for cumene hydroperoxide, particularly since more routes for by-product formation exist with cyclohexylbenzene hydroperoxide cleavage. Moreover, cyclohexanone is much more prone to acid-catalyzed aldol condensation reactions than acetone so that significant yield loss is possible unless the cyclohexylbenzene hydroperoxide cleavage is closely controlled.
There are other disadvantages of using sulfuric acid for cyclohexylbenzene hydroperoxide cleavage: 1) sulfuric acid is corrosive, especially in the presence of water, requiring expensive materials for reactor construction; 2) sulfuric acid needs to be neutralized before product separation and distillation, which requires additional chemicals such as phenate, caustics, or organic amines; and 3) the salt generated from neutralization requires separation and disposal and the waste water needs to be treated. Therefore, there are strong incentives to replace sulfuric acid with a heterogeneous cleavage catalyst that eliminates these drawbacks.
The patent and academic literature is replete with suggestions for replacing sulfuric acid in the cleavage of cumene hydroperoxide. For example, U.S. Pat. No. 4,490,565 discloses that zeolite beta is an effective replacement for sulfuric acid in the cleavage of cumene hydroperoxide and indicates that the yields, conversions and selectivities are generally superior to those produced by the use of the large pore zeolites X and Y. In U.S. Pat. No. 4,490,566, similar improvements over the large pore zeolites X and Y are reported with intermediate pore size zeolites, such as ZSM-5. In contrast, in an article entitled “Efficient Cleavage of Cumene Hydroperoxide over HUSY zeolites: The role of Bronsted activity”, Applied Catalysis A: General, 336 (2008), pages 29-34, Koltonov et al. report that cumene hydroperoxide readily undergoes decomposition over HUSY zeolites of high (15 to 40) Si/Al ratio with good selectivity to phenol and acetone and with efficiency even comparable to that of sulfuric acid. Despite or possibly because of these varying recommendations, most commercial processes for the cleavage of cumene hydroperoxide continue to use sulfuric acid as the catalyst.
Less interest has been focused on the cleavage of cyclohexylbenzene hydroperoxide, although International Patent Publication No. WO2011/001244 discloses that cyclohexylbenzene hydroperoxide can be converted to phenol and cyclohexanone in the presence of a variety of homogeneous or heterogeneous acid catalysts selected from Bronsted acids and Lewis acids. Suitable homogeneous catalysts are said to include protic acids selected from sulfuric acid, phosphoric acid, hydrochloric acid, and p-toluenesulfonic acid. Solid Bronsted acids such as Amberlyst and Lewis acids selected from ferric chloride, zinc chloride, boron trifluoride are also disclosed. In addition, suitable heterogeneous acids are said to include zeolite beta, zeolite Y, zeolite X, ZSM-5, ZSM-12 and mordenite.
Further, Japan Unexamined Patent Publication 2007-099746 discloses that cycloalkyl benzene hydroperoxides can be cleaved with high selectivity to phenol and cycloalkanone in the presence of montmorillonite, silica-alumina, cationic ion exchange resins, and sulfonic acid, perfluorosulfonic acid and heteropolyacids supported on a carrier. Similarly, Japan Unexamined Patent Publication 2007-099745 discloses that cycloalkyl benzene hydroperoxides can be cleaved with high selectivity to phenol and cycloalkanone in the presence of aluminosilicate zeolites having pore diameter of 0.6 nm or greater, such as zeolite Y and zeolite beta.
According to the present invention, it has now been surprisingly found that, although a number of solid acids, including aluminosilicate zeolites, have activity for the conversion of cyclohexylbenzene hydroperoxide into phenol and cyclohexanone, large pore zeolites of the FAU type having a unit cell size of less than 24.35 Å exhibit a unique combination of high activity and high selectivity for this reaction. Other large pore zeolites such as Beta do not give high selectivity. Moreover, given their high surface area, high silica FAU zeolites can potentially be used as a sorbent for removal and recovery of the N-hydroxyphthalimide (NHPI) catalyst used in the preceding oxidation step. Such NHPI removal could even be conducted concurrently with the cyclohexylbenzene hydroperoxide cleavage step, thereby simplifying the overall process.